Data Scientist

The Role
We're looking for an experienced AI Engineer to independently design, build, and scale LLM-powered and agentic AI systems for real-world enterprise use cases. You'll own projects end-to-end—from architecture and experimentation through deployment, evaluation, and observability—and help shape technical direction across the team.
Key Responsibilities
● Build and continuously improve LLM-powered applications for real-world enterprise use cases
● Design and develop agentic AI systems, including specialized agents, tool use, multi-step workflows, and MCP-based integrations
● Architect and implement RAG pipelines end-to-end: retrieval, re-ranking, context construction, and rigorous evaluation
● Build and integrate MCP servers and tools to connect AI agents with enterprise systems and data
sources
● Design and build scalable backend services and APIs to power AI applications in production
● Own evaluation of LLM applications—set up experimentation frameworks and drive both quantitative and qualitative assessment
● Build and maintain observability for AI systems—tracing, logging, dashboards, and alerting to monitor model/agent quality, latency, cost, and failures in production
● Work extensively with enterprise databases, vector databases, APIs, and internal data sources to build robust pipelines
● Partner closely with data scientists, AI/software engineers, and domain experts to take ideas from
prototype to production
● Optimize systems for latency, cost, and scalability; set up MLOps infrastructure including CI/CD,
monitoring, and model serving
● Mentor junior engineers and help raise technical standards across the team
Required Qualifications
● 3–6 years of experience in AI/ML engineering, applied data science, or a related field, with a track record of shipping LLM/AI applications to production
● Strong Python skills and solid software engineering practices (testing, CI/CD, code review, system
design)
● Proven, hands-on experience building LLM-based applications—RAG, agentic systems, or
tool-augmented workflows—at production scale
● Practical experience with MCP (Model Context Protocol) or comparable agent-tool integration
approaches
● Strong backend engineering skills—building and scaling APIs and services for AI applications
● Hands-on experience setting up evaluation frameworks (offline/online evals, A/B testing,
human-in-the-loop review) for LLM applications
● Experience with LLM observability tooling—tracing, logging, and monitoring for agent/model behavior, latency, cost, and quality regressions
● Experience working with enterprise/vector databases and integrating diverse internal data sources
● Solid grasp of ML/LLM fundamentals and the judgment to make architecture and trade-off decisions independently
● Experience with cloud platforms (AWS/GCP/Azure), containerization (Docker); Kubernetes a plus
● Bachelor's/Master's degree in Computer Science, Data Science, Engineering, or related field (or
equivalent practical experience)
● Strong communication skills and experience mentoring or leading technical work
Preferred / Nice-to-Have
● Experience with distributed training or large-scale data processing
● Hands-on experience with observability platforms (e.g., LangSmith, Langfuse, Arize, Datadog, Grafana) for AI/LLM systems
● Experience with RLHF or custom model alignment work
● Published research, patents, or significant open-source contributions in AI/ML
● Experience designing ML/AI platform infrastructure used by multiple teams.
Why Join Us?
This is an opportunity to shape the security and governance foundations of AI within one of the world's leading commodities trading companies. You will define how AI safely interacts with enterprise systems, ensuring innovation is delivered with the trust, transparency and controls expected in a highly regulated global business.

Work arrangement
No